Cystic fibrosis (CF) is a common, severe autosomal recessive disease caused by mutations in the CFTR gene. The CFTR gene encodes for a chloride channel responsible for chloride transport in epithelial cells. The major manifestations of CF are in the lungs, with more than 90% mortality related to the respiratory disease. The disease in the respiratory tract is linked to the insufficient CFTR function in the airway epithelium.
As of today, approximately 2000 different mutations disrupting the CFTR functions have been identified worldwide, grouped into five distinct classes based on their effect on the CFTR function (Rogan M. P. et al., 2011). Class I includes mutations that lead to non-functional CFTR (large deletions and stop codon mutations). Class II mutations (including the common F508del) lead to aberrantly folded CFTR protein that is recognized by the cell quality control mechanism and subsequently degraded, resulting in the absence of mature CFTR protein at the apical cell membrane. Class III mutations lead to full-length CFTR protein being incorporated into the cell membrane, but with defective regulation so that no CFTR function is present. These three classes usually lead to a classic CF phenotype with pancreatic insufficiency, although the severity of lung disease is highly variable. CFTR mutations leading to defective chloride conductance are grouped into Class IV. Class V mutations involve transcription dysregulation, resulting in a decreased amount of otherwise normal CFTR. The latter two classes are often associated with a milder phenotype and pancreatic sufficiency. Specifically, CFTR that results from a class IV mutation inserts into the plasma membrane but exhibits reduced single-channel chloride ion conductance because of reduced chloride permeation and open channel probability. R117H, among the most common class IV mutations, occurs at a worldwide frequency approaching 0.5%. The R117H missense mutation causes an arginine-to-histidine substitution at residue 117. R117H-CFTR R domain is normally phosphorylated, and the nucleotide binding domain (NBD) binds adenosine triphosphate (ATP), but channel open time and thus chloride transport are reduced. Additionally, the degree of R117H-CFTR function depends on the length of the polythymidine tract in intron 9 on the same chromosome (which influences splicing efficiency) such that the longer thymidine tracts (9T>7T>5T) produce more functional R117H-CFTR. Clinical disease typically requires the R117H mutation in cis with 5T (Rogan M. P. et al., 2011; Kiesewetter et al., 1993). Found in <1% of patients with CF, class V mutations produce normal plasma membrane CFTR. The quantity, however, is generally reduced as a result of transcriptional dysregulation. Class V mutations frequently influence the splicing machinery and generate both aberrantly and correctly spliced mRNA, the levels of which vary among different patients and even among different organs of the same patients. Ultimately, the splice variants result in a reduced number of functioning CFTR in the plasma membrane (Rogan M. P. et al., 2011).
About 10-15% of CFTR mutations affect the correct splicing of the gene transcripts. Among these are two mutations that are included in the invention: the first is the splicing mutation 3849+10 kb C-to-T which leads to inclusion of an 84 base pair cryptic exon in the mature messenger RNA (mRNA) (denoted “intron 22 cryptic exon inclusion” mutation). The mutation is the 12th most common CFTR mutation in the world, which occurs in hundreds of CF patients worldwide (Kerem et al., 1997; www.genet.sickkids.on.ca/; www.genet.sickkids.on.ca/resource/Table1.html). Correction of said aberrant splicing of the CFTR gene by “anti-sense” oligonucleotides was recently attempted by Friedman et al, 1999.
The second mutation is better described as a sequence variation in the poly (TG)n(T)n tract at the acceptor splice site of exon 10 affecting the retention of this exon in the mature mRNA (denoted “exon 10 exclusion” mutation). Importantly, the skipping of the exon results in a non-functional gene transcript, as the exon encodes for the first 21% of the intra-cytoplasmic nucleotide binding fold 1 (NBF1), a critical region for the CFTR function (Cutting et al., 1990; Kerem B. S. et al., 1990). The CFTR gene in many individuals, healthy or CF patients, has an inherent splicing inefficiency of exon 10 due to the non-optimal length of the sequence (TG)n(T)n with alleles carrying the (TG)13(T)5 combination generating the highest skipping levels (Chu et al., 2003; Hefferon et al., 2004; Groman et al., 2004).
One of the most promising therapeutic approaches for the treatment of genetic disorders caused by splicing mutations is based on splice-switching “anti-sense” oligonucleotides (AOs) administration. AOs are short synthetic RNA-like molecules chemically modified, which can anneal to motifs predicted to be involved in the pre-mRNA splicing. Their binding to selected sites is expected to mask the targeted region and promote normal splicing. AOs are highly specific for their targets and do not affect any other sequences in the cells. Several types of chemically modified AO molecules are commonly used including: 2′-O-methyl-phosphorothioate (2OMP), phosphorodiamidate morpholino oligomer (PMO), peptide nucleic acids (PNAs), 2-methoxyethyl phosphorothioate (MOE) and alternating locked nucleic acids (LNAs). Two of these are in more common use, 2OMP and PMO.
The AOs modifications maintain their stabilization, improve their target affinity, and provide favorable pharmacokinetic properties and biological stability. It has been conclusively shown that splice-switching AOs can redirect dystrophin pre-mRNA processing in murine models for Duchene Muscular Dystrophy (DMD) so that an exon carrying a premature protein termination signal (nonsense mutation) can be excluded from the mature gene transcript resulting in a shorter but still functional dystrophin isoform (Mann et al., 2001). Progress in dystrophin exon skipping has been rapid, with proof-of-concept studies reported in 2007 (van Deutekom et al., 2007) and 2009 (Kinali et al., 2009), and more recently with the publication of results from systemic administration to patients (Goemans et al., 2011; Cirak et al., 2011; Mendell J. R. et al., 2013). Systemic administration of OMP (5 weekly subcutaneous injections in 12 patients) showed dose-dependent molecular efficacy in patients with DMD (new dystrophin expression in muscle fibers), with a modest improvement in the 6-minute walk test (6MWT) in 8/10 patients which entered a 12 week extension study (Goemans et al., 2011). Systemic administration of PMO (AVI-4658) (12 weekly IV infusions) (Cirak et al., 2001) caused in 7/19 of the patients exon skipping and dystrophin restoration. Moreover, in a recent study published by Mendell J R et al. (Mendell J. R. et al., 2013) the ability of AVI-4658 to induce dystrophin production and to improve distance walk on the 6MWT was evaluated following 48 weeks of weekly IV infusions AVI-4658 restored functional dystrophin expression, causing a mean increase of 47% of dystrophin-positive fibers (change from baseline) together with an improvement in the 6MWT.
In addition to induced exon skipping, AOs can be designed to mask splice-silencing elements that reduce exon recognition and subsequent inclusion in the mature mRNA. Spinal Muscular Atrophy (SMA) is a common autosomal recessive condition (Lorson, Rindt, & Shababi, 2010) caused by the loss of the SMN1 gene together with a C>T variation in SMN2 exon 7, leading to abnormal splicing in which SMN2 exon 7 is skipped, resulting in a non-functional gene product. AOs have been designed to mask nearby flanking SMN2 splice silencer elements to promote synthesis of full-length transcripts (Singh, Shishimorova, Cao, Gangwani, & Singh, 2009; Mitrpant et al., 2009). An intrathecally administration of morpholino oligomer to neonatal mouse pups with severe SMA was highly successful, significantly extending their survival (Porensky et al., 2012).
Different routes of AOs delivery have been examined in animal models and applied in clinical trials, chosen primarily according to the target tissue. For example, 2OMP was administrated to DMD patients (PRO-051) by local intramuscular injection (van Deutekom et al., 2007), and by abdominal subcutaneous injections (Goemans et al., 2011). 2OMP was also administrated to a SMA mouse model by intracerebroventricular injection (Williams et al., 2009; Hua et al., 2010). PMO was administrated to a DMD mouse model by intramuscular injection (Gebski, Mann, Fletcher, & Wilton, 2003), and repeated weakly intraperitoneal injections (Goyenvalle et al., 2010). PMO was also administrated to a SMA mouse model by intracerebroventricular injection (Porensky et al., 2012), and to DMD patients (AVI-4658) by local intramuscular injection (Kinali et al., 2009), or intravenously administration (Cirak et al., 2011; Mendell et al., 2013).
There remains a constant need in the field of Cystic Fibrosis management for novel, potent therapeutics, designed to overcome the numerous mutations in the CFTR gene identified thus far, and restore CFTR function.